Six connected lines of work, from methodological development in statistical process monitoring and robust modelling to applied, data-driven research.
Research themes
i.
Profile monitoring and control charts
Phase I and Phase II monitoring of regression profiles with fixed and random effects, using parametric, nonparametric, semiparametric, Bayesian and multivariate charts.
A Semiparametric Mixed Model Approach to Phase I Profile Monitoring, Quality and Reliability Engineering International, 2013
Novel Bayesian CUSUM and EWMA control charts via various loss functions, Quality and Reliability Engineering International, 2023
Robust profile monitoring for phase II analysis via residuals, Quality and Reliability Engineering International, 2022
Economic capital and stress-test forecasting models from mortgage and business banking risk work at JPMorgan Chase (2011–2013), multi-step forecasting for VAR models, time series prediction for the Qatar stock market, and econometric studies of taxation and investment.
Assessing the Potential Impact of Introducing VAT on Price Levels in Qatar, International VAT Monitor, 2021
Stationary Bootstrap Based Multi-Step Forecasts for Unrestricted VAR Models, Journal of Data Science, 2020
On the Gaussian Process for Stationary and Non-stationary Time Series Prediction for the Qatar Stock Market, Springer, 2024
Machine-learning and deep-learning models for prediction and classification in education, health and the environment, benchmarked against classical statistical methods, including funded projects applying big data to Qatar Rail planning and machine learning to patient classification in emergency departments.
Performance prediction in online academic course: a deep learning approach with time series imaging, Multimedia Tools and Applications, 2023
Use of machine learning to assess factors affecting progression, retention, and graduation in first-year health professions students in Qatar, BMC Medical Education, 2023
On the Investigation of Monthly River Flow Generation Complexity Using the Applicability of Machine Learning Models, Complexity, 2021
Illustrations of EWMA and CUSUM control charts, profile monitoring, and robust versus least-squares regression, generated from simulated data to explain the ideas behind the research.